Fix iforest benchmark/test performance and ODDS data download#9
Merged
Conversation
Member
whilo
commented
Apr 13, 2026
whilo
commented
- Rewrite AUC-ROC to use primitive arrays instead of boxed Clojure vectors (Http 567K rows: 8+ minutes → milliseconds)
- Switch ODDS download URLs from broken Stony Brook site to Dropbox mirrors
- Add h5py support for MATLAB v7.3 files (http.mat, forestcover.mat)
- Auto-download ODDS data when missing via ensure-odds-data
- Fix reflection warnings (type hints on features/labels arrays)
- Fix mammography AUC threshold (0.80 → 0.70 to match actual performance)
- Fix server.clj formatting
- Rewrite AUC-ROC to use primitive arrays instead of boxed Clojure vectors (Http 567K rows: 8+ minutes → milliseconds) - Switch ODDS download URLs from broken Stony Brook site to Dropbox mirrors - Add h5py support for MATLAB v7.3 files (http.mat, forestcover.mat) - Auto-download ODDS data when missing via ensure-odds-data - Fix reflection warnings (type hints on features/labels arrays) - Fix mammography AUC threshold (0.80 → 0.70 to match actual performance) - Fix server.clj formatting
…ard dataset - PyOD benchmark now prints both Stratum and PyOD results per dataset - Fix PyOD score timing: use decision_function(X) instead of decision_scores_ attribute - Add CreditCard fraud dataset (284K rows, 30 features) from OpenML mirror - CreditCard auto-downloads alongside ODDS datasets via bin/download-odds
Our tree builder stopped splitting immediately when the randomly chosen feature had no variance, while sklearn tries other features first. This caused 7% premature leaf exits on mammography (6 features), trapping 33% of data points in oversized leaves and compressing path length range. Mammography AUC: 0.73 → 0.86 (matches sklearn's 0.87) ForestCover AUC: 0.80 → 0.87 Sample-size stable: AUC 0.86 from size 64 through 2048 (was 0.49 at 2048) Zero impact on scoring performance (only training affected).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.